Where to get historical intraday stock data?

External question — Quantitative Finance Stack Exchange Author: No Body Original post: https://quant.stackexchange.com/questions/51043 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. I hate to ask questions that have been asked before. But I am afraid that this is one of them. I have searched the web for days now, read so many forum posts. But I can't find an answer. Most answers about historical intraday data were about sites that offer historical futures and options, but I don't need those (btw. why would you want to use options data, which is derived from the stock data, instead of the actual stock data?). What I need is historical intraday stock data : At least S&P 500 stocks, dating back 5 years, 30 minute interval (or smaller), a weekly update, OHLC would be nice, but only one of them (close for example) would be sufficient. I need to be able to download the raw data, that means that I can't use something like Quantopian. So the data either needs to be downloadable (like a ZIP folder) or have a python API, so I can download it with the API. I know, something like this is not free, I would be able to spent up to about $50 a month . I am still a student, so I can't afford much more than that. What I found so far is Kibot and QuantQuote. Kibot : They seem to have a bad reputation. It seems that their stock data is either inaccurate, or incomplete. I just don't know if this still applies to the last 5 years. Since their data dates back to 1998, I could imagine that they had issues in the beginning but fixed them since they have been in business for more than 20 years. Their Standard subscription costs \$49 per month and includes 1 minute intraday data for all their symbols (6500 stocks, + ETFs, Futures and Forex) with a weekly update dating back 1 year. So I would also need to purchase once their historical S&P 500 data with a 30 minute interval, dating back to 1998 (I would need 2015 and up) for \$150. I can handle slightly inaccurate data (for example if the pricing is off by 0.5% or something like that), but missing data or inconsistent data (for example if instead of a constant 1min interval, it sometimes jumps multiple minutes) is a problem. QuantQuote : They are expensive, too expensive. I would need to purchase once their S&P 500 package for \$895 and then \$50/month for a weekly update. They seem to offer the same as Kibot, but for a much higher price and with a higher quality (I assume). It would help a lot, if you could say something about Kibot or Quantqote, or just about buying historical intraday data. Maybe you know better sites? Or do you know where to look or ask? Any comment is appreciated! EDIT: Well no one is answering or commenting, that's sad. I also found IQFeed . They have 10 years of minute data. They cost \$50 initially and after that about \$100 per month, plus \$130 for the program that downloads the data from their feed.

External answer — Quantitative Finance Stack Exchange Author: mirik Original post: https://quant.stackexchange.com/a/51055 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. There is a demo ticks dataset for S&P500: https://github.com/Jackal08/financial-data-structures/blob/master/raw_tick_data/ES_Trades.csv.zip It has 5.5kk entries for 20 days of the year 2013. The another source of intraday data is here: https://www.finam.ru/profile/akcii-usa-bats/google-inc/export/?market=25&em=20590&code=GOOG&apply=0&df=17&mf=10&yf=2019&from=17.11.2019&dt=17&mt=10&yt=2019&to=17.11.2019&p=7&f=GOOG_191117_191117&e=.txt&cn=GOOG&dtf=1&tmf=1&MSOR=1&mstime=on&mstimever=1&sep=1&sep2=1&datf=1&at=1 The site is in Russian but you can translate it with a Google translate.

External answer — Quantitative Finance Stack Exchange Author: Alex Norcliffe Original post: https://quant.stackexchange.com/a/51153 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. There are a wealth of providers out there. Your best bet is to sign up for Alpaca markets, which then gives you a free API key to use Polygon.io - they have 1-minute bar aggregates going back a decade for 11k US securities

External answer — Quantitative Finance Stack Exchange Author: ontic Original post: https://quant.stackexchange.com/a/51159 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. You asked why someone would want option data rather than stock data. Option price data is derived from the trading activity of people trading options. If you want to simulate an options trading strategy, you'll need option data. The data of the underlying stocks is not enough. You can't calculate the bid/ask/last trading price of an option based on the bid/ask/last of the underlying.

External answer — Quantitative Finance Stack Exchange Author: Judo Original post: https://quant.stackexchange.com/a/53675 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. You could also try firstratedata.com , which has downloadable csv files as opposed to an API. They have 1-min and ticks going back about 10-15 years. It seems they only update monthly, although you might need to check that. I tried Kibot previously and had no issues with the data, from the traders I know it doesn't have a bad reputation.

External answer — Quantitative Finance Stack Exchange Author: mightytravels Original post: https://quant.stackexchange.com/a/53811 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. eoddata.com has 1 minute bar data available for just USD 44.95 a month.

External answer — Quantitative Finance Stack Exchange Author: DataDaddy Original post: https://quant.stackexchange.com/a/60806 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. Yeah I would go with Alpaca Markets. I setup a small daily factor calculation using Alpaca Markets and a raspberry pi. Alpaca Markets also has a listener you can use to stream intra-day data.

External answer — Quantitative Finance Stack Exchange Author: louis_jones Original post: https://quant.stackexchange.com/a/85694 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. I bit the bullet and got mine from marketparquet.com. its like $279 for pretty much 1min bars for 15years. EODDATA only gives you 1 year and I kind of need to backtest couple of bear stretches.

External answer — Quantitative Finance Stack Exchange Author: wu jeff Original post: https://quant.stackexchange.com/a/85695 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. China’s second‑hand marketplace Xianyu, many sellers offer CSV files rather than APIs. These files usually include tick data plus pre‑compiled 1‑minute, 5‑minute, 15‑minute and other time‑frame data. I have personally purchased such data and found its quality to be excellent, at a cost of roughly 4 US dollars. By comparison, an API with 43‑millisecond latency costs around 60 RMB, or approximately 9 US dollars.